[S] Multiple failure time data

Badih.Ghattas (Badih.Ghattas@ehess.cnrs-mrs.fr)
Sat, 06 Jun 1998 09:47:56 +0200


Hi every body,

i'm dealing with data concerning 60 subjects each observed many times.
At each observation, a subject may be in two states 0 or 1(failure).
The number of observations and the number of failures are different for each
subject.
Also, the subjects are not observed at the same moment, and the time between
observations isn't regular.
Finally i have also covariates depending on time and on subjects.

I would like to model the failure time and to predict the status of a
subject at a given time according to the covariates' values.

My question :

-Are the multivariate failure time models (Kalbfleisch 7.3) the adequate
ones for such data ? I also heard about alternated renewal process and
points processes.
- If i want to fit a multivariate failure time data, can i do this with S+
for all the subjects, or do i have to fit a model per subject.

THANKS FOR YOUR HELP.

-------------------------------------------------------------------------------
Badih GHATTAS
GREQAM
Groupement de Recherche en Economètrie Quantitative d'Aix-Marseille
Centre de la Vieille Charité : 2, rue de la charité 13002 Marseille
Tel : (33) 04.91.14.07.35 FAX : (33) 04.91.90.02.27 ou
Tel : (33) 04.91.82.92.93 FAX : (33) 04.91.82.93.56
e-mail : ghattas@ehess.cnrs-mrs.fr
--------------------------------------------------------------------------------

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news